Package SMAP. R topics documented: June 19, 2018

Transcription

1 Package SMAP June 19, 2018 Title A Segmental Maximum A Posteriori Approach to Array-CGH Copy Number Profiling Version Date Depends R (>= 2.10), methods Author Robin Andersson <robin.andersson@lcb.uu.se> Functions and classes for DNA copy number profiling of array-cgh data Maintainer Robin Andersson <robin.andersson@lcb.uu.se> License GPL-2 LazyLoad true Collate AllClasses.R AllGenerics.R methods-smaphmm.r methods-smapobservations.r methods-smapprofile.r methods-gaussparam.r methods-grad.r methods-eta.r biocviews Microarray, TwoChannel, CopyNumberVariation git_url git_branch RELEASE_3_7 git_last_commit 3219eed git_last_commit_date Date/Publication R topics documented: GBM smap SMAPHMM SMAPHMM-class SMAPObservations SMAPObservations-class SMAPProfile-class SMAPProfiles-class Index 12 1

2 2 smap GBM Glioblastoma multiforme array CGH data Array CGH data measurements of glioblastoma multiforme sample G Usage data(gbm) Source Genome wide array CGH data from Diaz de Stahl, T., et al. (2005). Diaz de Stahl, T., et al. (2005) Chromosome 22 tiling-path array-cgh analysis identifies germline- and tumor-specific aberrations in patients with glioblastoma multiforme. Genes Chromosomes Cancer 44(2), smap Examples data(gbm) observations <- SMAPObservations(value=as.numeric(GBM[,2]), chromosome=as.character(gbm[,3]), startposition=as.numeric(gbm[,4]), endposition=as.numeric(gbm[,5]), name="g24460", reporterid=as.character(gbm[,1])) plot(observations) smap smap: A Segmental Maximum A Posteriori Approach to Array-CGH Copy Number Profiling This function fits a Hidden Markov Model (HMM) to a set of observed microarray intensity ratios and outputs the most plausible state sequence in the HMM through segmental a posteriori maximization. Briefly, given an HMM with initial parameter settings lambda and a set of observations O, the method alternates maximization of the joint posterior probability of the state sequence Q and lambda given O, p(q,lambda O), over Q (using a modified Viterbi algorithm) and lambda (using a gradient descent scheme with individual learning rate adaptation).

3 smap 3 Usage smap(x, Obs, sd.min=0.05, mean.sd=0.05, max.iters=inf, gd.max.iters=inf, tau=0.05, eta=0.01, e.change=0.5, e.same=1.2, e.min=0.0001, e.max=0.5, adaptive=true, overlap=true, distance=true, chrom.wise=false, verbose=1, L= ) Arguments x Obs sd.min mean.sd max.iters gd.max.iters tau eta e.change e.same e.min e.max adaptive overlap distance chrom.wise verbose L An object of class SMAPHMM-class. An object of class SMAPObservations-class. The minimum allowed standard deviation of state associated Gaussian distributions (numeric). Prior standard deviation of state associated Gaussian means (numeric). Maximum number of iterations in the SMAP algorithm (numeric). Maximum number of iterations in the gradient descent algorithm per SMAP iteration (numeric). Minimum log probability improvement required in the SMAP and gradient descent optimization (numeric). Initial learning rate in the gradient descent optimization (numeric). Multiplier for individual learning rate adaptation if the sign of partial derivative changes (numeric). Only used if adaptive == TRUE. Multiplier for individual learning rate adaptation if the sign of partial derivative stays the same (numeric). Only used if adaptive == TRUE. Minimum allowed learning rate (numeric). Maximum allowed learning rate (numeric). If TRUE, individual learning rate adaptation according to Algorithm 1 in Bagos et al. (2004) is used in the gradient descent optimization. If TRUE, genomic overlap of clones is considered in the optimization. If TRUE, genomic distance between clones is considered in the optimization, in terms of distance based transition probabilities. If TRUE, the observations are analyzed chromosome-wise rather than genomewise. Specifies the amount of output produced; 0 means no information and 3 a lot of information (numeric). A positive length parameter that controls the convergence of distance based transition probabilities towards 1 / nostates(x) (numeric). Details sd.min, mean.sd, and eta must all be greater than 0. tau must be greater than 0 if max.iters or gd.max.iters are infinite, and can be 0 otherwise. If adaptive is TRUE, then e.change is required to be in the interval (0,1], e.same must be greater than or equal to 1, and e.max must be greater than 0.

4 4 smap Value The method returns an object of class SMAPProfile-class or SMAPProfiles-class if chrom.wise is set to FALSE or TRUE, respectively. Robin Andersson Bagos P. G., Liakopoulos T. D., Hamodrakas, S. J. (2004) Faster Gradient Descent Training of Hidden Markov Models, Using Individual Learning Rate Adaptation. In Paliouras, G., Sakakibara, Y., editors, ICGI, volume 3264 of Lecture Notes in Computer Science, pages SMAPHMM, SMAPObservations Examples ## Load Glioblastoma multiforme data data(gbm) observations <- SMAPObservations(value=as.numeric(GBM[,2]), chromosome=as.character(gbm[,3]), startposition=as.numeric(gbm[,4]), endposition=as.numeric(gbm[,5]), name="g24460", reporterid=as.character(gbm[,1])) plot(observations, ylim=c(0,2)) ## Initiate HMM init.means <- c(0.4, 0.7, 1, 1.3, 1.6, 3) init.sds <- rep(0.1, 6) phi <- cbind(init.means, init.sds) hmm <- SMAPHMM(6, phi, inittrans=0.02) hmm ## RUN SMAP: profile <- smap(hmm, observations, verbose=2) ## genome profile plot(profile, ylim=c(0,2)) ## chromosome 9 profile ids <- which(chromosome(observations) == "9") plot(profile[ids], ylim=c(0,2), main="chromosome 9") ## output results for chromosome 9 #cbind(reporterid(observations[ids]), Q(profile[ids]))

5 SMAPHMM 5 SMAPHMM Constructor for "SMAPHMM" objects A constructor for SMAPHMM-class objects. Usage SMAPHMM(noStates, Phi, A=NULL, Pi=rep(1/noStates,noStates), inittrans=0.2/(nostates - 1)) Arguments nostates Phi A Pi inittrans The number of hidden states in the HMM (numeric). A Gaussian distribution parameter matrix (numeric). A nostates * nostates matrix of transition probabilities betweeen the hidden states (numeric). A vector of initial probabilities of starting in a certain state (numeric). Specifies the transition probability between non-equal states (numeric). Details Value Phi is a nostates * 2 matrix that specifies the paramaters of Gaussian distributions associated with each hidden state. The first column specifies standard deviations, the second specifies means. If A == NULL, inittrans specifies the transition probability between states i and j in 1:noStates, such that i!= j. Only used if A == NULL. inittrans * nostates must be smaller than (or equal to) 1. An object of class SMAPHMM-class. Robin Andersson, <robin.andersson@lcb.uu.se> smap, SMAPHMM-class, SMAPObservations-class

6 6 SMAPHMM-class SMAPHMM-class Class "SMAPHMM": A class to manage HMMs for the SMAP package Holds parameters for a Hidden Markov Model (HMM) used in the SMAP package. Objects from the Class Objects should not be created directly but via the constructor function SMAPHMM. Slots A: Object of class "matrix" The transition probability matrix between states. Pi: Object of class "numeric" The initial probabilities of starting in a certain state. Phi: Object of class "matrix" A matrix that specifies the paramaters of Gaussian distributions associated with each hidden state. The first column specifies standard deviations, the second specifies means. nostates: Object of class "numeric" The number of hidden states in the HMM. Z: Object of class "matrix" Matrix of transition probabilities. Y: Object of class "numeric" Vector of initial probabilities. eta: Object of class "ANY". Internal slot. grad: Object of class "ANY". Internal slot. Methods A signature(object = "SMAPHMM"): Returns the transition matrix. Pi signature(object = "SMAPHMM"): Returns the initial probabilities. Phi signature(object = "SMAPHMM"): Returns the distribution parameter matrix. nostates signature(object = "SMAPHMM"): Returns the number of hidden states in the HMM. Robin Andersson, <robin.andersson@lcb.uu.se> smap, SMAPHMM

7 SMAPObservations 7 SMAPObservations Constructor for "SMAPObservations" objects A constructor for SMAPObservations-class objects. Usage SMAPObservations(value, chromosome, startposition, endposition, name=character(0), reporterid=as.character(1:length(value))) Arguments value chromosome startposition endposition name reporterid A vector of microarray intensity ratios (numeric). A vector of chromosome annotations (character). A vector of start positions (numeric). A vector of end positions (numeric). The name of the observation set (character). A vector of observation identifiers, e.g., probe ids (character). Details The vectors value, chromosome, startposition, endposition, and reporterid must be of equal length. Value An object of class SMAPObservations-class. Robin Andersson, <robin.andersson@lcb.uu.se> smap, SMAPObservations-class, SMAPHMM-class

8 8 SMAPObservations-class Examples ## Load Glioblastoma multiforme data data(gbm) observations <- SMAPObservations(value=as.numeric(GBM[,2]), chromosome=as.character(gbm[,3]), startposition=as.numeric(gbm[,4]), endposition=as.numeric(gbm[,5]), name="g24460", reporterid=as.character(gbm[,1])) ## plot observations plot(observations, ylim=c(0,2)) ## plot subset of observations (chromosome 9) ids <- which(chromosome(observations) == "9") plot(observations[ids]) SMAPObservations-class Class "SMAPObservations": A class to manage microarray observations for the SMAP package Holds observed microarray intensity ratios and clone annotations for the SMAP package. Objects from the Class Objects can be created by calls of the form new("smapobservations", value, chromosome, startposition,endpos Values for internal slots (see below) are not intended to be passed upon construction. You can also use the convenience function SMAPObservations. Slots value: Object of class "numeric" Microarray intensity ratios. chromosome: Object of class "character" Associated chromosomes for the observations. startposition: Object of class "numeric" Associated start positions for the observations. endposition: Object of class "numeric" Associated end positions for the observations. reporterid: Object of class "character" Identifiers of the observations, e.g., probe ids. name: Object of class "character" An identifier of the observation set. noobservations: Object of class "numeric" The number of observations in the set. chrom.start: Object of class "numeric". Internal slot. chroms: Object of class "character". Internal slot. distance: Object of class "numeric". Internal slot. nooverlaps: Object of class "numeric". Internal slot. overlaps: Object of class "numeric". Internal slot. overlapids: Object of class "numeric". Internal slot. startoverlaps: Object of class "numeric". Internal slot.

9 SMAPProfile-class 9 Methods value signature(object = "SMAPObservations"): Returns the values of the observations. chromosome signature(object = "SMAPObservations"): Returns the chromosome annotations of the observations. startposition signature(object = "SMAPObservations"): Returns the start positions of the observations. endposition signature(object = "SMAPObservations"): Returns the end positions of the observations. reporterid signature(object = "SMAPObservations"): Returns the identifiers of the observations. name signature(object = "SMAPObservations"): Returns the name of the observation set. noobservations signature(object = "SMAPObservations"): Returns the number of observations in the set. initialize signature(.object = "SMAPObservations"): Creates an instance. plot signature(x = "SMAPObservations", y = "missing"): A plot method for the observations. [ signature(x = "SMAPPObservations"): Creates a new object of class SMAPObservations with extracted elements as specified by the indices provided. Robin Andersson, <robin.andersson@lcb.uu.se> smap, SMAPObservations SMAPProfile-class Class "SMAPProfile" Holds results from running smap. Objects from the Class Objects are not intended to be created directly but as a result from running smap.

10 10 SMAPProfiles-class Slots HMM: Object of class "SMAPHMM" observations: Object of class "SMAPObservations" P: Object of class "numeric" The log joint posterior probability of the state sequence Q and parameters of HMM given the observations. Q: Object of class "numeric" The optimal state sequance (path) in the HMM. name: Object of class "character" The name of the object. Methods P signature(object = "SMAPProfile"): Returns the log joint posterior probability. Q signature(object = "SMAPProfile"): Returns the optimal state sequence. HMM signature(object = "SMAPProfile"): Returns the optimized HMM. name signature(object = "SMAPProfile"): Returns the name of the profile. observations signature(object = "SMAPProfile"): Returns the observations. plot signature(x = "SMAPProfile", y = "missing"): A plot method for the result profile. [ signature(x = "SMAPProfile"): Creates a new object of class SMAPProfile with extracted elements as specified by the indices provided. Robin Andersson, <robin.andersson@lcb.uu.se> smap, SMAPProfiles-class SMAPProfiles-class Class "SMAPProfiles" Holds results from running smap. Objects from the Class Objects are not intended to be created directly but as a result from running smap. Slots.Data: Object of class "list" A list of objects of class SMAPProfile-class. name: Object of class "character" The name of the object.

11 SMAPProfiles-class 11 Extends Class "list", from data part. Class "vector", by class "list". Methods Q signature(object = "SMAPProfiles"): Returns the optimal state sequence of the list elements. observations signature(object = "SMAPProfiles"): Returns the observations of the list elements. name signature(object = "SMAPProfiles"): Returns the name of the profile. plot signature(x = "SMAPProfiles", y = "missing"): A plot method for the result profiles. Robin Andersson, <robin.andersson@lcb.uu.se> smap, SMAPProfile-class

12 Index Topic classes SMAPHMM, 5 SMAPHMM-class, 6 SMAPObservations, 7 SMAPObservations-class, 8 SMAPProfile-class, 9 SMAPProfiles-class, 10 Topic datasets GBM, 2 Topic methods smap, 2 startposition (SMAPObservations-class), 8 value (SMAPObservations-class), 8 A (SMAPHMM-class), 6 chromosome (SMAPObservations-class), 8 endposition (SMAPObservations-class), 8 GBM, 2 HMM (SMAPProfile-class), 9 name (SMAPObservations-class), 8 noobservations (SMAPObservations-class), 8 nostates (SMAPHMM-class), 6 observations (SMAPProfile-class), 9 P (SMAPProfile-class), 9 Phi (SMAPHMM-class), 6 Pi (SMAPHMM-class), 6 plot,smapobservations,missing-method (SMAPObservations-class), 8 Q (SMAPProfile-class), 9 reporterid (SMAPObservations-class), 8 smap, 2, 2, 5 7, 9 11 SMAPHMM, 4, 5, 6 SMAPHMM-class, 6 SMAPObservations, 4, 7, 8, 9 SMAPObservations-class, 8 SMAPProfile-class, 9 SMAPProfiles-class, 10 12